Together AI introduces ThunderAgent – doubling AI agent speed
Together AI introduces ThunderAgent, a program-aware scheduler that doubles throughput for agent-based inference during synthetic data generation.
What happened?
Together AI has announced ThunderAgent, a new scheduler for AI agents that optimises execution during synthetic data generation. By treating each agent workflow as a schedulable program, the system eliminates KV-cache thrashing. This enables more than double the throughput on individual nodes and near-linear scalability across multiple nodes.
Key facts
| Genomströmning enskild nod | >2x ökning |
|---|---|
| Skalbarhet över flera noder | Nästan linjär |
| Huvudfunktion | Eliminerar KV-cache-thrashing |
”By treating each agent workflow as a schedulable program, it eliminates KV cache thrashing to deliver more than 2x single-node throughput and near-linear multi-node scaling.”
Why it matters
Efficient generation of synthetic data is a critical bottleneck for training next-generation AI models. By preventing unnecessary flushing of the KV-cache, hardware costs are reduced significantly while the speed of complex agent workflows is doubled.
Who is affected?
The technology is primarily aimed at AI researchers, developers, and companies building large-scale agent-based systems and generating synthetic datasets for model training.
Impact on the EU
Not relevant to EU status. The source code and system architecture are focused entirely on infrastructural performance and scheduling.
What else you should know
The system demonstrates how program-aware scheduling can maximise the utilisation of computing resources in synthetic data generation. Tests show that it avoids the bottlenecks that have previously limited large-scale parallel execution of AI agents.
Quick answers about this story
Vad har hänt?
När hände det?
Varför spelar det roll?
Vilka påverkas av ThunderAgent?
The link opens in a new window and leads to the publisher's own site.
Källan har spårats automatiskt från utgivaren via Aheadlines signalkedja.
AI-verktyg i artikeln
Topics
Get similar news straight to your inbox
The reader's room
Send in a question or an addition. The newsroom reads everything before it's published and replies when relevant. No AI-generated text – just people.
Sign in to submit a comment or question.
Read the article through your role
- Assess technical risk: model choice, vendor lock-in, data flow and running cost.
- Update the architecture doc if new APIs or regulations touch production.
- Ensure observability + rollback plan before rolling out to production.
Generated angle — not editorial analysis of "Together AI introduces ThunderAgent – doubling AI agent spee"